Journal
/
/
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
JoVE Journal
Engineering
This content is Free Access.
JoVE Journal Engineering
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
DOI:

04:48 min

November 30, 2022

, , ,

Chapters

  • 00:04Introduction
  • 00:42Eyeball, Optic Nerve, and Extraocular Muscle Masking on the Orbital CT Scans
  • 01:42Pre-Processing: Window Clipping and Cropping the VOIs
  • 02:16Four Cross-Validations of the Orbital Segmentation Model
  • 02:57Results: Segmentation Results of the Orbital Structures
  • 04:01Conclusion

Summary

Automatic Translation

An object segmentation protocol for orbital computed tomography (CT) images is introduced. The methods of labeling the ground truth of orbital structures by using super-resolution, extracting the volume of interest from CT images, and modeling multi-label segmentation using 2D sequential U-Net for orbital CT images are explained for supervised learning.

Related Videos

Read Article